Tolerability and efficacy of duloxetine in a nontrial situation
JRA Duckett, Consultant Gynaecologist, Medway Maritime Hospital, Windmill Road, Gillingham, Kent ME7 5NY, UK. Email firstname.lastname@example.org
Objective To assess the tolerability and efficacy of duloxetine in a nontrial situation.
Design Prospective observational study.
Setting Urogynaecology Unit, District General Hospital, UK.
Population Two hundred and twenty-two women with a diagnosis of urodynamic stress incontinence (USI) or mixed USI and detrusor overactivity (DOA) took duloxetine for 4 weeks.
Methods The results of therapy were assessed with a Patient Global Impression of Improvement (PGI-I) questionnaire. One hundred and forty-eight (67%) women were initially treated with 40 mg twice a day, 67 (30%) women were treated with an escalating dose initially at 20 mg twice a day increasing to 40 mg twice a day after 2 weeks and seven (3%) women were started on a dose of 20 mg twice a day which they continued.
Main outcome measures Discontinuation rates and PGI-I scores.
Results Overall 146/222 (66%) women discontinued therapy due to adverse effects or lack of efficacy. Significantly more women starting on the 40 mg twice a day dose stopped due to adverse effects when compared with the escalating dose (P < 0.025). Of the women who tolerated therapy, 80 out of 120 (67%) had a PGI-I score indicating an improvement. However, the overall rate of improvement was 37%. PGI-I scores and discontinuation rates were not significantly different between the group with USI and the group with mixed USI and DOA (P > 0.05).
Conclusion In a nontrial situation duloxetine is poorly tolerated. Introducing an escalating dose may improve tolerability. A similar number of women with USI and mixed incontinence had a PGI-I score indicating improvement.
The involuntary leakage of urine1 (urinary incontinence [UI]) is common among women.2 UI presents as stress urinary incontinence (SUI), mixed urinary incontinence or urge urinary incontinence.1 SUI is defined as the complaint of involuntary leakage of urine upon effort or exertion, or on sneezing or coughing.1 SUI is the most prevalent type of UI in younger and middle-aged women3 and is known to significantly affect quality of life (QoL).4 SUI can be treated conservatively by means of pelvic floor muscle training and lifestyle interventions such as weight loss, regulating fluid intake and smoking cessation.5 In the past the only treatment option if conservative therapy failed was surgery. Duloxetine is the first drug to be licensed for use in the treatment of SUI. It is a potent and relatively balanced serotonin (5H-T) and noradrenaline (NA) reuptake inhibitor. Pooled data analysis from randomised, double-blind placebo-controlled trials on duloxetine, has shown duloxetine to be significantly more effective than placebo in reducing incontinence episode frequency (IEF) and improving QoL scores.6–9
Inhibition of 5H-T and NA reuptake is not only the basis of duloxetine’s efficacy but also the cause of its adverse effects. The most common adverse effect is nausea and was reported to have a prevalence of 23% in the above trials.6–9 Nausea, dizziness, fatigue and insomnia (1–6%) were the most common adverse event-related reasons for discontinuation. The revised guidelines of the International Consultation on Incontinence recommend complementing conservative therapy with appropriate drug therapy for the initial management of women with SUI.10 Duloxetine was introduced in the UK for the treatment of moderate/severe incontinence in 2004 and therefore this was incorporated into our unit treatment algorithms. There is currently very little data relating to the use of duloxetine outside clinical trials sponsored by the manufacturer. The study was designed to assess the acceptability and tolerability of duloxetine and compare this with the original trials.
Materials and methods
All women presenting to the urogynaecology unit were assessed and treated at a one-stop clinic, on the basis of a full structured history and examination, urinalysis, Kings QoL questionnaire,4 frequency volume chart and cystometry. Women with a diagnosis of urodynamic stress incontinence (USI) were offered pelvic floor exercises (PFEs), duloxetine or surgery depending on their previous treatment. PFE and duloxetine were recommended to women if they had not previously been treated with conservative therapies. Those women previously treated with PFE or anticholinergic drugs were also offered duloxetine and a repeat course of PFE. Physiotherapy treatment was not started in any woman prior to evaluation of the effect of the initial 4 weeks of duloxetine treatment because of the limited immediate availability of nurse practitioners and physiotherapists. Those women tolerating duloxetine were treated with concurrent physiotherapy when this was available. Hence, all women in this cohort were primarily treated solely with duloxetine for 4 weeks. Data were collected on all women receiving duloxetine from the launch of the drug in November 2004 until September 2006. Women not assessed with cystometry were excluded from the study.
Before starting treatment, the patients were advised of the common adverse effects of therapy and advised that the medication would probably be required in the longer term. Women who declined drug therapy were excluded from further analysis. Our previous work has suggested that approximately 40% of women decline drug therapy and choose surgical intervention once advised of the adverse effect profile and lack of long term data.11 Written literature provided by the makers of duloxetine was used to inform women. The five commonest adverse effects were discussed: nausea, fatigue, insomnia, dizziness and somnolence.12 Women were advised that adverse effects would resolve in 80% of cases if the drug was taken for more than 4 weeks. They were advised that there was at least a 50% chance that they would have an improvement in their incontinence. The counselling was not altered when the escalating dose was introduced. A lack of long-term data was discussed. Patients were given a 4-week course of duloxetine. A 4-week period was used as 100% response has been demonstrated by day 15 in previous studies13 and adverse effects are likely to resolve by this point.
Women with mixed USI and detrusor overactivity (DOA) were considered suitable for treatment if they were predominantly complaining of moderate/severe stress incontinence. These patients were offered the same treatment as described for women with USI alone, because data exists which suggest there is a role for duloxetine treatment in women with mixed incontinence.12
Duloxetine was initially prescribed at a dose of 40 mg twice daily as per the manufacturer’s recommendations. In 2005, an abstract presented at the annual International Continence Society meeting suggested using an escalating dose of 20 mg twice a day for 2 weeks and then increasing to 40 mg twice a day in order to reduce the incidence and severity of adverse effects.14 The treatment schedule offered to our patients was therefore changed to this regimen. Efficacy and tolerability outcomes were calculated only using women who completed the study having taken duloxetine 40 mg twice a day for at least 2 weeks. A small number of women continued to take 20 mg twice a day at the assessment visit. These women were excluded as this dose has failed to show an improvement in QoL assessments in previous studies.
Women were graded according to the patient Global Impression of Severity (PGI-S) and Improvement questionnaires (PGI-I).15 These are validated questionnaires that have previously been used in clinical trials in subjects receiving duloxetine and placebo. The PGI-S scale allows women to rate their incontinence as mild, moderate or severe and the PGI-I scale allows women to rate their improvement after treatment on a seven-point scale from ‘Very much better’ to ‘Very much worse’, with the centre of the scale as ‘No change’ (see Table 3 for details). Women scoring 1–3 on this scale were considered to show an improvement for the purposes of analysis. Reasons for discontinuation of therapy were documented, e.g. lack of efficacy, adverse effects. Specific adverse effects were not recorded as these have been well documented in previous trials of duloxetine.
Table 3. Improvement reported by those women taking duloxetine for 4 weeks
|Very much better||8 (16)||10 (14)|
|Much better||11 (22)||22 (32)|
|A little better||14 (28)||15 (22)|
|No change||10 (20)||21 (30)|
|A little worse||3 (6)||1 (1)|
|Much worse||3 (6)||1 (1)|
|Very much worse||1 (2)||0 (0)|
Results were entered into a database and analysed using SPSS version 13.0 (Chicago, IL, USA). All data entries were checked by a second researcher. Chi-squared tests were used to analyse data, with a P value of less than or equal to 0.05 considered significant. Differences in percentages are presented with the 95% CI of the difference. Specific ethical committee approval was not sought for this study. Duloxetine was used as per the manufacturer’s licence and the data collected and outcome measures were part of the usual assessment of the incontinent woman within our department protocols. All terminology conforms to the International Continence Society standards.1
Two hundred and twenty-two women attending the urogynaecology unit with a diagnosis of USI chose to take duloxetine. One hundred and nine women recorded that they had previously been treated with PFEs. The patient demographics are detailed in Table 1. Forty-two women with mixed incontinence had not previously received any anticholinergic medication. Among the remaining 57 women, 30 had received tolterodine tartrate XL 4 mg, 24 trospium chloride 20 mg twice a day, seven solifenacin succinate 5 mg daily, two propiverine hydrochloride 20 mg twice a day and one oxybutynin 2.5 mg three times a day. Seven women were treated with two or more anticholinergics prior to the instigation of duloxetine therapy. The urodynamic study had been performed when the woman was not taking anticholinergic drugs. The six women taking anticholinergics and duloxetine together, the anticholinergic had been taken for a minimum of 6 weeks before duloxetine was started.
Table 1. Patient demographic data
|Number of women||99||123|| |
|Mean Age||52.2 (±5.1)||52.6 (±9.7)||0.64|
|Previous anticholinergic drugs|
|At any time||57 (58)||0|| |
|Still receiving||6 (6)||0|| |
|Mild||5 (5)||13 (11)|| |
|Moderate||45 (45)||70 (57)||<0.05|
|Severe||49 (50)||40 (32)|| |
|0-6||4 (4)||26 (21)|| |
|7-14||28 (28)||31 (25)||<0.001|
|>14||67 (68)||66 (54)|| |
The first 148 women (67%) were treated with 40 mg twice a day and the remaining 67 women (30%) were treated with an escalating dose, starting treatment at a dose of 20 mg twice daily for 2 weeks before increasing to the recommended dose of 40 mg twice daily. Seven women (3%) started and continued on the 20 mg twice a day regimen rather than increasing to 40 mg twice a day (although we did not recommend this treatment option). Four women decreased their dose from 40 mg twice a day to 20 mg twice a day and stayed on that dose after the initial assessment at 4 weeks.
Overall, 146/222 (68%) women discontinued therapy due to adverse effects or lack of efficacy (Table 2). Significantly more women starting on the 40 mg twice a day dose stopped due to adverse effects when compared with the escalating dose (50% versus 31%; difference 18.7% [95% CI 4.3; 33.0]) With the escalating regimen a similar number of women discontinued due to lack of efficacy on the standard regimen as the escalating regimen (25% versus 21%; difference 4.1% [95% CI −8.2; 16.4]) when compared with the standard regimen. There were no serious adverse events.
Table 2. Discontinuation rates on duloxetine
|All||146 (68)||111 (75)||35 (52)||22.8 (9.3; 36.2)||<0.001|
|Adverse effects (%)||95 (65)||74 (50)||21 (31)||18.7 (4.3; 33.0)||<0.02|
|Lack of efficacy (%)||51 (35)||37 (25)||14 (21)||4.1 (−8.2; 16.4)||NS|
There was no statistically significant difference between the discontinuation rates due to adverse effects in the USI and mixed groups (P > 0.05). Forty-three out of 99 (43%) women in the mixed group discontinued due to adverse effects and 52 out of 123 (42%) women in the USI group discontinued due to adverse effects.
Of the women who tolerated therapy, 80 out of 120 (who completed the assessment on a dose of 40 mg twice a day) (67%) had a PGI-I score indicating an improvement (Table 3). The proportion of women reporting a PGI-I score of 1–3 was not significantly different between the group with mixed USI and DOA and the group with USI alone (66% versus 67.1%; difference 1.1% (95% CI −16.0; 18.3). A similar number of women in each group were ‘much better’ or ‘very much better’ (38% mixed; 45.7% USI; difference 7.7% (95% CI −10.2; 25.7).
Nine women recorded worsening incontinence (PGI-I scores of 5–7). Seven of the 9 women had mixed incontinence. Four women completing the 4-week course of treatment preferred surgery rather than long-term medication despite improvement in their PGI-I score. Despite stopping the drug due to insufficient benefit 14 women reported an improvement in the PGI-I (all scored 3). If the PGI-I scores of women who could not tolerate duloxetine are included, the overall improvement rate (PGI-I scores of 1–3) after initiating treatment with duloxetine was 37% (80/215).
The discontinuation rate for duloxetine was higher than suggested by previous randomised controlled trials. At the recommended dose of 40 mg twice a day, duloxetine was poorly tolerated compared with the placebo-controlled trials. In those women who tolerated therapy the efficacy was less than previously suggested.
Randomised controlled trials are considered the best method for providing information on safety and efficacy and they provide good evidence to bring a drug to the market. They give less information about how the drug may be received outside the context of a placebo-controlled trial. Placebo-controlled trials have been criticised because they focus on highly selected populations and outcomes.16 Duloxetine trials suggest benefit in the treatment arm of between 49–65% and low discontinuation rates due to adverse effects (22%).6–9 A Cochrane review reports that duloxetine significantly improved the QoL of patients with SUI.17 The review reported that one in eight women stopped duloxetine due to adverse effects.
In women who did not stop due to adverse effects 80/120 (67%) had a PGI-I score indicating improvement. However, if the results are recalculated including those women who stopped due to adverse effects, only 37% had a PGI-I score indicating improvement. This was worse than those reported in the pooled analysis from the Phase 3 randomised placebo-controlled studies on a worldwide scale, which was 75%.6–9 Interestingly a similar number of women with mixed incontinence and USI had a PGI-I score indicating improvement. This suggests that there may be a role for duloxetine in the treatment of women with mixed incontinence. A recent study has also reported a significant improvement in the IEF in women with mixed incontinence. It showed a 63% improvement in IEF for women with USI and a 65% improvement of IEF in women with mixed incontinence.12
The discontinuation rate using an escalating regimen of duloxetine was less than when the drug was introduced at the recommended dose of 40 mg twice a day. Comparing different doses of duloxetine was not an initial aim of the study and should be interpreted with caution outside the context of a randomised study, designed specifically to look at this issue. There are no obvious reasons why this observation should not be true and the two populations were well matched with no differences in demographics or counselling (data not shown). It is difficult to identify any confounding factor which could have resulted in such a large difference in tolerability between the two dosing regimens and therefore the escalating dose is likely to improve tolerability.
Physiotherapy treatment was not started in any woman prior to evaluation of the effect of duloxetine treatment because of the limited immediate availability of nurse practitioners and physiotherapists. Ideally, a one-stop service including a consultation, diagnosis and combined treatment should operate, thus ensuring that both treatments are started together.
A small number of women (predominantly in the mixed group) reported a worsening of their incontinence. A further analysis will be performed to see if any factors predict which women deteriorated. This might provide the opportunity to make recommendations about women who are not suitable to try duloxetine. It is interesting to speculate why the women in this study were less able to tolerate duloxetine compared with those of the previous trials. The improved tolerability seen in clinical trials is possibly due to the ‘supportive network’ which is generally available for patients who are undergoing a clinical trial. The process of questioning and assessments may help women understand their condition and take action to control their symptoms. This study also raises the question about the validity of commercially sponsored placebo-controlled trials in assessing the tolerability, adverse effects and efficacy of drug trials in the treatment of incontinence. Further research should also concentrate on the long-term outcomes of drug therapy as this was only a short trial and success rates/continuation rates may well fall with time.
Women taking duloxetine at the recommended dose are less likely to tolerate therapy than suggested by the original papers. An escalating dose may be used to improve tolerability. Women with mixed DOA and USI are equally likely to show an improvement than women with USI alone.